Real Time Computer Vision

This SBIR Phase II project will develop and demonstrate efficient background separation technologies that enable real-time computer vision in mobile sensors and devices. The Phase II will build upon the enabling technologies and the prototype software developed in Phase I that use advanced mathematical models and subspace estimation algorithms, which are several orders-of-magnitude faster than the state-of-the-art methods. The advantage of these technologies is that they do not require any hardware optimization or special computational architecture to achieve real-time performance and can be executed on a variety of mobile processors. Furthermore, as opposed to many state-of-the-art approaches, they enable background separation on videos from freely moving sensors in addition to static sensors. The Phase II effort will include improving the core technologies, optimizing and porting the technologies for efficient processing to handle high-resolution video data, quantitative analysis of detection software, and demonstration of the technologies in realistic environments. The project will benefit from Novateur Teams expertise in the areas of background subtraction, automated visual surveillance, online matrix completion, and adaptable sensor exploitation.